ZMF Caldera Headphone Review

[Benesyed]

If people could choose between a Caldera without EQ and one EQed to Harman, a large majority would prefer the Harman one. This research has been done and there are no serious doubts about this. You always make the same argument that preferences vary and so on, but it is simply not true, you either don't know or acknowledge sufficiently the research. This is also largely independent of age, origin, gender and training. That many people buy your phones says nothing about their sound quality, Beats by Dr. Dre were a massive success as well ( I guess a tiny bit larger success even).

I'm not sure the exact FRC of the Caldera was compared to Harman. I would need to see what was actually compared and how.

Would be interesting to see who preferred what.

Tbh it would also be good experimental design to have can't tell the difference/no strong preference as an option. And maybe have 3 headphones, 2 the same, one different to actually assess if people could reliably tell things apart...

 

[ReaderZ]

Our drivers are proprietary and designed from the ground up and built for us from approved vendors who follow our needs. The most important aspects of this driver are the unique magnet shape for the acoustic properties I was after, the membrane material and plating/coating techniques and materials used. I wanted to make sure our MFR OEM didn't make planar drivers for any other brand, which can be tough. Unfortunately I can't share suppliers, but we utilize numerous suppliers for everything we do so to meet the specs and performance I'm after as well as tolerances.

Thanks for the detailed answer! Waiting for my stabilized VC now.

 

[MacClintock]

There's plenty of headphones out there that adhere to Harman, that are under $500 - why would anyone buy a Harman tuned headphone like the DCA stuff that's way more expensive given the tuned is the same? AKG 361/371 is a great example. Why buy anything more expensive if FR is all that matters?

Ah, no. I am not sure if I can take you seriously. The AKG 361 for example has a preference rating of 76 and after EQ to Harman 98, according to the oratory measurements. So not exactly Harman. The AKG 371 has 85 before EQ and 99 after. So a bit bitter but still quite off. But I will help you a bit, the Sennheiser HD600 has a rating of 91 before EQ, so is really quite good, but has very low bass.
Sure for headphones there is more than frequency reponse (FR), but as the Harman research has shown, FR determines preference by a large percentage.
So people buy probably the expensive DCA headphones, because they don't need EQ to have good sound quality (and nice design and premium materials, etc).
But buying a $3,500 headphone that still needs EQ and does nothing better than for example the Audeze LCD-X is only for people who are either not well informed, or want to show off.

 

[Benesyed]

Ah, no. I am not sure if I can take you seriously. The AKG 361 for example has a preference rating of 76 and after EQ to Harman 98, according to the oratory measurements. So not exactly Harman. The AKG 371 has 85 before EQ and 99 after. So a bit bitter but still quite off. But I will help you a bit, the Sennheiser HD600 has a rating of 91 before EQ, so is really quite good, but has very low bass.
Sure for headphones there is more than frequency reponse (FR), but as the Harman research has shown, FR determines preference by a large percentage.
So people buy probably the expensive DCA headphones, because they don't need EQ to have good sound quality (and nice design and premium materials, etc).
But buying a $3,500 headphone that still needs EQ and does nothing better than for example the Audeze LCD-X is only for people who are either not well informed, or want to show off.

Or comfort/build related issues too. The DCA expanse is not my most comfortable headphone for my head sadly.

The best for that metric is the much maligned MDR Z1R. Those are amazing comfort wise.

 

[zach915m]

Ah, no. I am not sure if I can take you seriously. The AKG 361 for example has a preference rating of 76 and after EQ to Harman 98, according to the oratory measurements. So not exactly Harman. The AKG 371 has 85 before EQ and 99 after. So a bit bitter but still quite off. But I will help you a bit, the Sennheiser HD600 has a rating of 91 before EQ, so is really quite good, but has very low base.
Sure for headphones there is more than frequency reponse (FR), but as the Harman research has shown, FR determines preference by a large percentage.
So people buy probably the expensive DCA headphones, because they don't need EQ to have good sound quality (and nice design and premium materials, etc).
But buying a $3,500 headphone that still needs EQ and does nothing better than for example the Audeze LCD-X is only for people who are either not well informed, or want to show off.

Oh - Oratory is great - I've met him and we really align on how we think about things. He's very pragmatic and open minded. A true scientist who understands the research and loves what he does.

Have you read this? https://www.reddit.com/r/oratory1990/comments/jl661e
I think it might apply in this situation a bit, here's an excerpt:

"Anything you hear, you can measure. But just because you can measure a difference does not mean you can hear a difference.
The question is: How to interpret a measurement.
"Detail" or "grain" aren't some magical, unmeasurable qualities that only exist in sound when the sound enters a human eardrum, but suddenly stop existing when the same sound enters a microphone.
The question is just, which aspects of the sound do we look at in order to put a number on the experience that we are getting.
But for this to be possible, first we have to agree on what we are hearing, when we say things like "detail".
And if you've ever been part of audiophile discussions you will see how hard this is:
People will quickly agree when talking about the same headphone at the same time. But ask them about their opinion without telling them the opinion of other audiophiles, and suddenly the results will be very different. Even worse: Don't tell them anything about the headphone they're listening to (blind-test), and the results will often not even be consistent."

---------------

Also since you're using the preference system, the DCA Aeon 2 Closed is a 91. The Stealth is an 86. Quite a difference in price. And I'm serious when I ask this becacuse I really would like to know an answer. If FR is all that matters, why would a company like DCA that is so steadfast to the curve have headphones tunes to the same curve in difference price ranges if there's not more to it than the FR? Why would the more expensive one be lesser on the preference curve?

 

[amirm]

If FR is all that matters, why would a company like DCA that is so steadfast to the curve have headphones tunes to the same curve in difference price ranges if there's not more to it than the FR?

Frequency response is the most important aspect of a headphone from technical point of view. Purchase criteria however goes beyond that. Looks, brand, fit, how hot they wear, cost, support, level of seal, etc. all come into play. You know this so don't know why you are asking.

 

[amirm]

Adding on, on the technical front, spatial qualities matter to some people (although many make up things about imaging, etc.). Distortion can also be an issue especially if you are going to EQ. Sensitivity or ease of driving is also important.

In other words, there are plenty of ways to innovate beyond frequency response.

 

[MacClintock]

Oh - Oratory is great - I've met him and we really align on how we think about things.

I have my doubts on the last assessment.

Also since you're using the preference system, the DCA Aeon 2 Closed is a 91. The Stealth is an 86.

92 and 88. It seems you have trouble in reading numbers properly.

If FR is all that matters,

Why do you keep repeating this all over again? Everybody, inluding Amir, told you that nobody is claiming this for headphones. But you are repeating it time and again to fight against ghosts. It seems you don't want to understand.

 

[amirm]

Oh - Oratory is great - I've met him and we really align on how we think about things. He's very pragmatic and open minded. A true scientist who understands the research and loves what he does.

I don't know him as a "scientist." An engineer, yes, but not scientist.

Even worse: Don't tell them anything about the headphone they're listening to (blind-test), and the results will often not even be consistent."

Indeed. Many audiophiles make poor listeners as they don't even agree with themselves no matter what lofty claims they make about their "experience." Again, from Harman research:

See where Audio Reviewers landed. I have a video on this that goes into the research:

I get to it at 30 minute mark but I highly suggest you watch the whole thing if you value science.

 

[IAtaman]

Yes agreed! And the bias would assume that the Harman Target would be the only way to tune a headphone to sound good, or with a touch more, or a touch less bass. My only argument in this thread and going forward based on my non-studied experience, is that there's other factors that can make a headphone sound better in specific form factor scenarios, tuned slightly away from Harman (yes more or less bass/treble). Some headphones need strict adherence to harman and some don't based on form factor. It's what I have experienced and I know many users who only trust research are turned off by that, and I get it. I could definitely have added in distortion using other methods to get a harman like tuning in the bass of the Caldera, I found it didn't work for my and many of our owners tastes. I'm about to release a headphone that's around 1K that is pretty closely adhered to harman in some pad/mesh formats, it's not that I don't know how to tune to harman.

But to assume bias for users who buy our headphones after hearing other harman headphones back to back (I've seen it many times) assumes that they are buying them based on just aesthetics or something, or because I'm a nice guy and some marketing genius, that's just not true. There's stuff that hasn't been studied yet, some of it was pointed out earlier by Mad Economist, and I firmly believe an open mind about audio, taking into account how great the harman curve is, will lead to better future results for us all.

You can safely ignore these keyboard warriors who think think they know everything because they have seen an FR graph.

Of course tuning is not all there is to it when it comes to the sound quality of a headphone. There are many other factors that contribute to the sound quality which we currently can not tell from a single FR graph. Spatial qualities is a good example - we can make an educated guess maybe based on the FR but we can not tell how a headphone will perform in this regard until we put it on our heads. Detail is another example and there might be others as well.

Tonality however is a different story. An FR graph will allow us to predict how a headphone might sound tonal balance wise.

An yes of course, Harman Target is not the only way to tune a headphone. It is what sounds natural to majority of the people regardless of age, background, nationality etc however and therefore it is generally preferable to the majority of the people, when neutrality is an objective. But it is also true that majority is not anonymity, and neutrality is not an objective always, and not everyone hears the tonality as it is measured by a measurement rig.

Given all these variations, what I think is important personally is that headphone manufacturers do recognize Harman target for what it is - a tuning that is what people prefer when they are looking for neutrality, and should be able to articulate tuning decisions that deviate from it with sensible reasoning.

Specifically for his headphone, due to where Amir chose to normalize, I believe the FR graph does not fully reflect how it would sound to most people timbre wise. I read quite a few reviews and impressions online, most people find it a bit bright. Normalizing at 700Hz indeed seems to be a better choice. If you do that, the hole in 1.5K would disappear, and 2K and 5K will overshoot a bit - not unlike the graph you have previously shared. And then there is the variation from the bass. Could you by any chance tell us why you made those tuning decisions you made?

 

[IAtaman]

But buying a $3,500 headphone that still needs EQ and does nothing better than for example the Audeze LCD-X is only for people who are either not well informed, or want to show off.

Oh nice. I see you made progress and can now tell why people make the purchasing decisions they make by looking at FR graphs. I suspect the day you would be able to tell what the designer had for lunch when they designed the headphone by looking at a FR graph is not too far away.

 

[MacClintock]

Oh nice. I see you made progress and can now tell why people make the purchasing decisions they made looking at FR graphs. I suspect the day you would be able to tell what the designer had for lunch the day they designed the headphone by looking at a FR is not too far away.

What a quirky remark, keyboard warrior.

 

[zach915m]

And then there is the variation from the bass. Could you by any chance tell us why you made those tuning decisions you made?

For the bass on an open headphone ways I found to get to harman hurt the sound and the impulse response making the transients of the headphone worse, causing the listener to perceive the headphone as more "compressed," which basically describes a worse impulse response and worse distortion. The Atrium Damping system "catches" air behind the driver with variable damping based on the system to alleviate for the natural linear bass response of a planar driver giving a higher impulse response and less THD to try to make up for some of the non-humped bass that is done via other methods. The lower THD allows EQ if desired which I prefer for planars over a pre-existing higher THD in the bass. Harman is certainly viable, and I hope anyone reading this thread knows that I believe in it, but that I have views one different headphone formats and what works best.

I personally tune all my headphones a little dark in the 3 to 5 KHZ are as that's my preference and I've found the preference of a lot of people who seem to like ZMF's, the treble can be tuned down at the 7-8 KHZ overshoot, I typically am not as sensitive in this area, and like a little sizzle there above the vocal and sibilance region, but I understand some users aren't so we make that mesh to accommodate for that.

 

[zach915m]

I have my doubts on the last assessment.

I really did meet Oratory and I do like him very much.

 

[MacClintock]

I really did meet Oratory and I do like him very much.

I am not questioning this, please read again what I have posted.

 

[zach915m]

I am not questioning this, please read again what I have posted.

I hope anyone reading this thread knows that I believe in it (Harman Target), but that I have views on different headphone formats and what works best.

 

[MacClintock]

Of course tuning is not all there is to it when it comes to the sound quality of a headphone.

And nobody here claimed this.

There are many other factors that contribute to the sound quality which we currently can not tell from a single FR graph. Spatial qualities is a good example - we can make an educated guess maybe based on the FR but we can not tell how a headphone will perform in this regard until we put it on our heads.

It is very doubtfull that spatial qualities can be explained from FR alone (or even to a large extent), which can be easily shown by EQing for example the HD800 to Harman and observing that it's spatial qualities still remain there.

Detail is another example and there might be others as well.

Detail is a myth which is nothing but high treble. Any headphone that can reproduce 20kHz is cabable of reproducing every possible detail there is in the source material.

 

[Madlop26]

We were making great progress encouraging Zach to do the blind test, would be way more productive than all these passive aggressive talk.

 

[solderdude]

You need to stop assuming stuff about ZMF. I'm an open book for most pragmatic things.

Our drivers are also not copies of the Foster drivers and if you had them in hand it would be very obvious,

I don't have one in hand and never have, I have to go on pictures like this:
Am just an amateur with an interest in headphones and not a professional nor YT reviewer.
On the left Denon (made by Foster) in the middle Atrium driver (made by ZMF), on the right Ali-Express 'copy'.

Based on that info one could conclude the drivers are either from the same company or they are optically close copies.
Since you are so secretive (contrary to being an open book) about driver manufacturers... what am I to think ?

And yes, most people do buy your headphones because of the woodwork and craftsmen ship (which is excellent).
They buy them because of reviews or forum posts or trying them.
Where is the misinformation about that ?

And yes, one can EQ most headphones to a certain target (but may still not sound the same)

For one thing I agree with many of your viewpoints regarding targets and preference of people buying headphones.
The vast majority of headphone owners will not EQ anyway and would not know how to.

If anything I kind of defended your products, my only gripe would be the price but as you are making a living and do not sell huge quantities and have to invest (why 2 BK5128 ?) I can understand where the price has to come from.

They appear to be made to high standards and don't think the sound of your models is anything but agreeable to your target audience.
I might even like the sound of them anyway but I do not know as they are too expensive for me.

 

[solderdude]

the DCA Aeon 2 Closed is a 91. The Stealth is an 86. Quite a difference in price. And I'm serious when I ask this becacuse I really would like to know an answer.

It is based on:
A: a single measurement trace (which could be an averaged one) with a certain resolution
B: an algo determines and weighs some number aspects of that trace.
C: only FR plot on a specific fixture is considered in the number generation
D: seal issues are not included
E: Fit and room for the ears is not considered

I can go on and think rating headphones on a single number is silly and not constructive. People like lists though so that's why they exist... for the 'list people'.
And to be fair to Amir, he does not rank headphones he just vents his opinion on them based on listening and measurements (acc to the standard he uses)
Ranking on preference of one person or majority voting (fans in a forum agreeing about a product they like) or using sale numbers is equally silly.

Ranking is silly and is just a list. I ignore all those lists. One should not think of these lists as 'pure science' but can use some of the additional data (if one understands what the numbers mean) but just not about how you will like that headphone on your head with your favorite recording/artist.

I think it is a bit silly to use the 'ranking' as an argument against Harman research or ASR.